JP2013251757A - Image processor and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve image quality of a boundary region between a background image and an object image.SOLUTION: An image processor includes: an image data acquisition section for acquiring target image data showing a target image including an object image and a background image adjacent to the object image; a feature value acquisition section for acquiring a first feature value on a color of the object image and a second feature value on a color of the background image; and a correction section for correcting the color of a boundary region being a region where the object image and the background image are adjacent to an intermediate color showing a color between the color of the object image and the color of the background image by using the first feature value and the second feature value.

Description

  The present invention relates to a technique for converting the color of a boundary region between an object image and a background image included in a target image.

  A technique for removing a background color (background color) of a target image represented by image data is known. For example, if the target image from which the background color is removed is printed, the amount of printing material (ink, toner, etc.) used can be reduced compared to the case of printing the target image before removal. is there.

  For example, in Patent Document 1, in the background color removal processing, an image area having a gradation value within the removal range is removed (converted to white), and the gradation value within the adjustment range that is continuous to the removal range is corrected. Thus, a technique for maintaining the continuity of gradation in an image after removal processing is disclosed. As a result, it is possible to alleviate the gradation step that occurs in the boundary area between the background image from which the background color has been removed and the object image.

JP 2001-94804 A

  However, depending on the target image, there is room for improving the image quality of the boundary area between the background image and the object image, and improvement in the image quality of the boundary area is required.

  The main advantage of the present invention is to improve the image quality of the boundary region between the background image and the object image in the image represented by the image data.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented in the following aspects.

Application Example 1 An image data acquisition unit that acquires target image data representing a target image including an object image and a background image adjacent to the object image, a first feature value relating to the color of the object image, The object image and the background image are adjacent using a feature value acquisition unit that acquires a second feature value relating to the color of the background image, and the first feature value and the second feature value. An image processing apparatus comprising: a correction unit that corrects a color of a boundary area, which is an area to be corrected, to an intermediate color indicating a color between the color of the object image and the color of the background image.

  According to the above configuration, the color of the boundary area between the object image and the background image is corrected to an intermediate color between the color of the object image and the color of the background image using the first feature value and the second feature value. . As a result, it is possible to improve the image quality of the boundary region by suppressing unnatural color that may occur in the boundary region.

Application Example 2 An image data acquisition unit that acquires target image data representing a target image including an object image and a background image adjacent to the object image, a first feature value relating to the color of the object image, A feature value acquisition unit that acquires at least one feature value of the second feature value relating to the color of the background image, and at least one feature of the acquired first feature value and the second feature value And the luminance value of the boundary area that is an area where the object image and the background image are adjacent to each other, the color of the boundary area is set to at least one of the color of the object image and the color of the background image An image processing apparatus comprising: a correction unit that corrects the color so as to approach the color of the image.

  According to the above configuration, the color of the boundary region is determined using the feature value of at least one of the first feature value and the second feature value and the luminance value of the boundary region between the object image and the background image. Correction is performed so as to approach at least one of the color of the object image and the color of the background image. As a result, color unnaturalness that may occur in the boundary region can be suppressed, and the image quality of the boundary region can be improved.

  The present invention can be realized in various forms, such as an image processing system, an image processing method, a computer program for realizing these functions, a recording medium on which the computer program is recorded, and the like. It can be realized in the form.

2 is a block diagram showing a configuration of a multifunction machine 200. FIG. It is a flowchart of an image process. It is a figure which shows an example of object image data. It is a figure which shows an example of the color conversion table used for a component color conversion process. It is a figure which shows the object image data after a background color conversion process was performed. It is a figure which shows the color on the straight line AA shown to FIG. 3 (B) and FIG. 5 (B). It is a flowchart of a color correction process. It is a figure which shows an example of a color correction table. It is a figure which shows an example of a color correction table.

A. Example:
A-1. Multi-function machine configuration:
Next, embodiments of the present invention will be described based on examples. FIG. 1 is a block diagram illustrating a configuration of a multifunction machine 200 as an image processing apparatus according to an embodiment.

  The multifunction device 200 includes a CPU 210, a nonvolatile storage device 220 such as a hard disk drive or an EEPROM, a volatile storage device 230 such as a RAM, and a printer unit 240 that prints an image by a predetermined method (for example, inkjet or laser). , A scanner unit 250 that reads a document using a photoelectric conversion element (for example, CCD, CMOS), an operation unit 260 such as a touch panel and buttons, a display unit 270 such as a liquid crystal panel superimposed on the touch panel, a digital camera 300, A communication unit 280 for performing data communication with an external device such as a personal computer 400 or a USB memory (not shown).

  The volatile storage device 230 is provided with a buffer area 231 for temporarily storing various intermediate data generated when the CPU 210 performs processing. The nonvolatile storage device 220 stores a computer program 222 for controlling the multifunction device 200. The computer program 222 can be provided in a form recorded on a CD-ROM, for example.

  The CPU 210 executes the computer program 222 to function as an image processing unit 100 that executes image processing to be described later, and a device control unit 20 that controls the printer unit 240 and the scanner unit 250. The image processing unit 100 includes an image data acquisition unit 110, a conversion unit 120, a feature value acquisition unit 130, a determination unit 140, a specification unit 150, and a correction unit 160.

A-2. Image processing:
The image processing unit 100 performs image processing on an object image representing an object such as a character, a photograph, or a figure, and image data representing a background representing the background of the object image. This image processing includes processing for converting the color of the background image and processing for correcting the color of the boundary area, which is an area near the boundary where the object image and the background image are adjacent to each other. This image processing is executed on image data to be printed prior to printing, for example, when a print instruction for printing an image represented by the image data using the printer unit 240 is received from the user. This print instruction may be a print job transmitted from an external device (for example, the personal computer 400 (FIG. 1)), or image data (scan data) is generated by reading a document using the scanner unit 250. Alternatively, it may be a copy instruction instructing printing using the scan data.

  The image data (target image data) to be processed in the image processing of the present embodiment is image data generated by an image generation device such as the scanner unit 250 or the digital camera 300 (FIG. 1), for example. In the image processing of this embodiment, scan data generated by reading an original including an object (for example, a character) often expressed in a single color by using an image reading device such as the scanner unit 250 is converted into target image data. As an example.

FIG. 2 is a flowchart of image processing.
In step S10, the image processing unit 100 acquires target image data. For example, when image processing is started in response to receiving a print job, the image data acquisition unit 110 includes image data included in the print job (for example, scan data stored in the personal computer 400). Are acquired as target image data. When image processing is started in response to receiving a copy instruction, the image data acquisition unit 110 generates scan data by reading a document prepared by a user using the scanner unit 250. The scan data is acquired as target image data. The target image data is, for example, RGB image data composed of pixel data including RGB component values (for example, 256 gradations).

  FIG. 3 is a diagram illustrating an example of target image data. FIG. 3A is an overall view of the target image 50 represented by the target image data. The target image 50 includes four object images, that is, a first character image 52 representing the character string “SAMPLE”, a second character image 53 representing the character string “EFGH”, and a third character image representing the character string “WXY”. A character image 54 and a photographic image 57 are included as partial images. Each of the three character images 52, 53, and 54 is a monochromatic image substantially represented by a single color, and the photographic image 57 is a multicolor image including a plurality of colors.

  Here, the term “substantially monochromatic” means that, for example, when an observer observes an object image at a normal observation distance, the object represented in the object image is recognized as monochromatic. This includes that the object is expressed in the selected color. For example, a character image that is substantially monochromatic is a color that is close enough to recognize that the character is monochromatic when viewed at an observation distance when reading the character. , Including that the character is represented. Further, the object image representing the object included in the scan data obtained by reading the document including the object determined to be substantially monochrome according to the above-described standard with an image reading device such as a scanner is further compared with the document. Thus, color differences caused by reading variations can occur. The object image included in the scan data is included in an object image that represents an object that is substantially monochrome.

  Furthermore, the target image 50 includes three background images, that is, a first background image 55 representing the background of the character image 52, a second background image 51 representing the background of the second character image 53 and the photographic image 57, and a third image. A third background image 56 representing the background of the character image 54 is included as a partial image. The background image is a partial image adjacent to the object image so as to surround all or part of the periphery of the object image. In general, a background image is an image that is substantially monochromatic and does not represent a specific object (characters, figures, photographs, etc.), and is also referred to as a background image. The second background image 51 is white, and the first background image 55 and the third background image 56 have a color different from white.

  FIG. 3B shows a schematic enlarged view in which a part of the first character image 52 (near the character L) is enlarged. As shown in FIG. 3B, the boundary region MG located at the boundary between the first character image 52 and the first background image 55 adjacent to the first character image 52 is the same as the color of the first character image 52. In some cases, the background image 55 has a color different from that of the background image 55. For example, the intermediate color of the boundary region MG is generated due to reading characteristics such as the resolution of the image reading apparatus, the reading speed, and the characteristics of the image sensor. For example, a general image reading apparatus generates image data by receiving light from a document by the image sensor while moving the position of the image sensor relative to the document in the sub-scanning direction. At this time, the pixel data of the boundary region MG can be generated based on both the light from the character and the light from the background. As a result, the color of the boundary region MG may have a color that is generated by mixing the character color of the document with the background color. For example, when the color of the first character image 52 is black and the color of the first background image 55 is relatively light red, the color of the boundary region MG may include a relatively dark red.

  In step S20, the conversion unit 120 performs a background color conversion process on the target image data. The background color conversion process of the present embodiment executes the color conversion process on the target image data, and changes the color of the background image to a specific color (white (R, G, B) = (255, 255, 255)).

  Since each pixel data of the target image data includes three component values (RGB values) constituting the RGB color system, it can be said that the target image data includes three component image data. . The component image data is image data composed of one component value among the three component values. The background color conversion process includes three component color conversion processes executed independently of each other. That is, the conversion unit 120 independently executes component color conversion processing using a color conversion table for each of the three component image data. Here, the component color conversion processing is executed independently, in order to convert the pixel value (for example, R value) of one component image data, the pixel value (for example, G value) of other component image data. Or B value).

  FIG. 4 is a diagram illustrating an example of a color conversion table used for component color conversion processing. This color conversion table RT corresponds to the change of the input gradation value Vin within the change range CR (TH1 <Vin ≦ 255) from the first threshold TH1 to the maximum gradation value (255). Vout is all changed to the maximum gradation value (255). The color conversion table RT has the same output gradation value Vout corresponding to the change of the input gradation value Vin within the maintenance range SR (0 ≦ Vin <TH2) from the minimum value (0) to the second threshold value TH2. Maintain the gradation value. Further, the color conversion table RT has an input gradation value Vin from the second threshold value TH2 to the first threshold value TH1 in the adjustment range MR (TH2 ≦ Vin ≦ TH1) from the second threshold value TH2 to the first threshold value TH1. The output gradation value Vout is set so that the output gradation value Vout continuously changes from the second threshold value TH2 to the maximum gradation value (255) corresponding to the change up to. Accordingly, the color conversion table RT is configured so that the output gradation value Vout continuously changes with respect to the change of the input gradation value Vin in the entire gradation range including before and after the thresholds TH1 and TH2. Is set.

The color conversion table RT shown in FIG.
Vout = Vin, (0 ≦ Vin <TH2)
Vout = {(255−TH2) / (TH1−TH2)} × (Vin−TH2) + TH2, (TH2 ≦ Vin ≦ TH1)
Vout = 255, (TH1 <Vin ≦ 255)
It can be expressed as.

  The component color conversion process is executed for all pixel values of the processing target component image data (processing target component values of all pixels included in the target image 50) using the color conversion table RT. Each component conversion process may be executed using a common color conversion table RT, or may be executed using a color conversion table RT determined for each component (RGB). The color conversion table RT for each component may have different threshold values TH1 and TH2, for example.

  Here, like a typical document, each object image (FIG. 3) of the target image 50 is represented by a relatively dark color (a color in which each RGB value is a relatively small value), and each background image is It is assumed that the color is expressed by a relatively light color (a color in which each RGB value is a relatively large value). The background color conversion process assumes image data representing an image such as the target image 50 as a processing target. For this reason, the thresholds TH1 and TH2 of the color conversion table RT are larger than the RGB values assumed as the pixel values of the object image and smaller than the RGB values assumed as the pixel values of the background image. Is set. For example, the first threshold value TH1 is set to 210, and the second threshold value TH2 is set to 190. As a result, the background color conversion process can convert the color of the background image included in the image represented by the image data assumed as the processing target into a specific color. Such background color conversion processing is executed to convert the color of the background image to white when printing an image where the color of the background image is not important, such as a character document (so-called background removal). Processing and ground removal processing). As a result, if an image is printed using the image data after the background color conversion process, the amount of printing material (toner or ink) required for printing can be reduced.

  FIG. 5 is a diagram illustrating the target image data after the background color conversion process is executed. Here, the target image data after the background color conversion process is executed is also called converted image data, and the target image data before the background color conversion process is executed is also called original image data. FIG. 5A is an overall view of the converted image 60 represented by the converted image data. Here, “6” is used for the second digit of the code of the partial image included in the converted image 60, and the code of the corresponding partial image in the original image 50 shown in FIG. Use the same number as the first digit. As the name of the partial image included in the converted image 60, the same name as the corresponding partial image in the original image 50 is used. For example, the first character image 62 and the second character image 63 of the converted image 60 correspond to the first character image 52 and the second character image 53 of the original image 50, respectively (FIGS. 3 and 5).

  The four object images (character images 62, 63, 64, and photographic image 67) of the converted image 60 are maintained in color without changing all the values of the three components (RGB) to 255 by the background color conversion process. ing. However, the object image may include pixels whose component values have been adjusted using the adjustment range MR of the color conversion table RT. The color of the first background image 65 is changed to white as a result of all the values of the three components being set to 255 by the background color conversion process. As for the color of the third background image 66, the values of some of the three components (for example, the R value and the G value) are changed to 255 by the background color conversion process, and the remaining values (for example, the B value) are maintained. As a result, it has been changed to a color different from white. Background images that have undergone substantial color change, such as the first background image 65 and the third background image 66, are referred to as changed background images. The color of the second background image 61 is not substantially changed by the background color conversion process, and is kept white. However, when the second background image 51 of the target image 50 includes unevenness (color variation), the unevenness is removed from the second background image 61 of the converted image 60 by the color conversion process. Like the second background image 61, a background image that is maintained in white, that is, a background image that has not been changed in color before and after the background color conversion process is also referred to as a maintenance background image.

  FIG. 5B shows a schematic enlarged view in which a part of the first character image 62 (in the vicinity of the character L) is enlarged as in FIG. 3B. The color of the boundary region MG located at the boundary between the first character image 62 and the first background image 65 is also maintained by the background color conversion process. An observer who has seen the converted image 60 may feel the color of the boundary region MG unnatural. For example, as described above, the color of the boundary region MG may have a color generated by a color mixture of the character color of the document and the background color. In the original image 50 described above, an example is considered in which the color of the first character image 52 is black, the color of the first background image 55 is relatively light red, and the color of the boundary region MG includes relatively dark red. . In this case, in the original image 50, the deep red color of the boundary region MG is a color related to both the black color of the first character image 52 and the light red color of the first background image 55. Inconspicuous. In the original image 50, the deep red color of the boundary region MG hardly gives an unnatural impression to the observer, but may give a natural impression. However, in the converted image 60, the deep red color of the boundary region MG is not easily related to the color (white color) of the first background image 65 after color conversion, so that it tends to stand out and give an unnatural impression.

  FIG. 6 is a diagram showing values obtained by converting the colors (RGB values) on the line AA shown in FIGS. 3B and 5B into the YCbCr color system. In the graph of FIG. 6A, the solid line YC1 indicates the luminance value Y of the original image 50, and the broken line YC2 indicates the luminance value Y of the converted image 60. In the graph of FIG. 6B, the solid line RC1 indicates the chroma value Cr of the original image 50, and the broken line RC2 indicates the chroma value Cr of the converted image 60. Dashed lines YC2 and RC2 show only portions that are different between the converted image 60 and the original image 50 (before and after the background color conversion process). Of the luminance value Y and chroma value Cr of the converted image 60, the luminance value Y and chroma value Cr of portions not shown by the broken lines YC2 and RC2 are the same as the luminance value of the original image 50 indicated by the solid lines YC1 and RC1. . As will be described later, a one-dot broken line RC3 indicates a chroma value Cr after a color correction process described later is performed on the converted image 60.

  The color values (also referred to as YCC values) representing the color of the object image (for example, the first character image 52) of the original image 50 in the YCbCr color system are referred to as (TC_Y, TC_Cb, TC_Cr). The color of the object image (for example, the first character image 62) of the converted image 60 is the same as the color of the object image of the original image 50. The YCC value representing the color of the background image (for example, the first background image 55) of the original image 50 is (BC_Y1, BC_Cb1, BC_Cr1). The YCC value representing the color of the background image (for example, the first background image 65) of the converted image 60 is (BC_Y2, BC_Cb2, BC_Cr2).

  As shown in FIG. 6A, the luminance value Y of the original image 50 is substantially constant TC_Y in the first character image 52 that is substantially monochromatic, and the first background image 55 that is substantially monochromatic. Is a substantially constant value BC_Y1. Actually, the luminance value Y of the first character image 52 and the first background image 55 includes fluctuation due to variation, but in FIG. 6A, it is shown as a constant value in order to avoid complication of the drawing. The luminance value Y of the boundary region MG is substantially the same as the luminance value TC_Y of the first character image 52 at the first position P1 in contact with the first character image 52, and at the second position P2 in contact with the first background image 55. It is almost the same as the luminance value BC_Y1 of the first background image 55. Then, at a position between the first position P1 and the second position P2, the luminance value Y of the boundary region MG is closer to TC_Y as the position is closer to the first position P1, and to BC_Y1 as the position is closer to the second position P2. There is a tendency to change to be closer. The tendency of this change is not necessarily linear due to variations or the like, but in FIG. 6A, it is shown as a straight line in order to avoid the complexity of the figure.

  Further, the luminance value Y of the converted image 60 is the same as that of the original image 50 in the first character image 52, and becomes a constant value BC_Y2 in the first background image 55. Since BC_Y2 is a white luminance value, it is larger than BC_Y1. The luminance value Y of the boundary region MG of the converted image 60 changes from the value in the original image 50 so that the luminance value Y does not become discontinuous in the vicinity of the second position P2. This is because the pixel value is adjusted by the adjustment range MR of the color conversion table RT (FIG. 4) of the background color conversion process. The luminance value Y of the boundary region MG of the converted image 60 is the same as the luminance value Y of the boundary region MG of the original image 50 in a range excluding the vicinity of the second position P2. As a whole, the luminance value Y of the boundary region MG of the converted image 60 is the same as the luminance value Y of the boundary region MG of the original image 50 from the first position P1 toward the second position P2. Is closer to TC_Y, and closer to BC_Y2 is closer to the second position P2.

  Thus, the tendency of change in the luminance value Y of the boundary region MG is also maintained in the converted image 60. Also, the background color conversion processing of the present embodiment is performed when the luminance value Y of the object image is relatively low (dark color) and the luminance value Y of the background image is relatively high (light color). Is a process of further increasing the luminance value Y of (for example, converting to white). For this reason, the change amount of the luminance value Y by the background color conversion process is relatively small. The luminance value Y generally does not cause the boundary region MG of the converted image 60 to give an unnatural impression.

  As shown in FIG. 6B, the chroma value Cr of the original image 50 is a substantially constant value TC_Cr in the first character image 52, and is a substantially constant value BC_Cr1 in the first background image 55. Actually, like the luminance value Y, the chroma values Cr of the first character image 52 and the first background image 55 include fluctuations due to variations, but are shown as constant values in FIG. 6B. The chroma value Cr of the boundary region MG of the original image 50 is substantially the same as the luminance value TC_Cr of the first character image 52 at the first position P1, and is substantially the luminance value BC_Cr1 of the first background image 55 at the second position P2. Is the same. At a position between the first position P1 and the second position P2, the chroma value Cr of the boundary region MG is closer to TC_Cr as it is closer to the first position P1, and closer to BC_Cr1 as it is closer to the second position P2. As such, there is a tendency to change. The tendency of this change is not necessarily linear due to variations or the like, but in FIG. 6B, it is shown as a straight line in order to avoid complication of the figure.

  The chroma value Cr of the converted image 60 is the same as that of the original image 50 in the first character image 62, and becomes a constant value BC_Cr2 in the first background image 65. Since the color of the first background image 65 is white in the present embodiment, BC_Cr2 is “0”. Therefore, for example, when the color of the first background image 55 of the original image 50 is a color with relatively high saturation (when BC_Cr1 is a relatively large value), the chroma value Cr is obtained by the background color conversion process. Can be seen to change greatly (FIG. 6B). The chroma value Cr in the boundary region MG of the converted image 60 changes from the value in the original image 50 so that the chroma value Cr does not become discontinuous in the vicinity of the second position P2. This is because the pixel value is adjusted by the adjustment range MR of the color conversion table RT (FIG. 4) of the background color conversion process. Then, the chroma value Cr of the boundary region MG of the converted image 60 is the same as the chroma value Cr of the boundary region MG of the original image 50 in a range excluding the vicinity of the second position P2. As a result, as in the example of FIG. 6B, when the chroma value Cr changes greatly due to the background color conversion process, the above-described tendency of the change of the chroma value Cr in the boundary region MG may be disrupted. I understand. For example, in the example of FIG. 6B, the chroma value Cr of the boundary region MG of the converted image 60 is not closer to TC_Cr as it is closer to the first position P1, but closer to BC_Cr2 as it is closer to the second position P2. It can be seen that a maximum value exists between the first position P1 and the second position P2. Such a change can occur in the other chroma value Cb as well as the chroma value Cr. Due to the characteristics of the chroma values Cr and Cb, the color of the boundary region MG of the converted image 60 may include an unnatural color. Hereinafter, a series of processes for correcting an unnatural color that may occur in the color of the boundary region MG of the converted image 60 will be described.

  Returning to FIG. Following the background color conversion process, in step S30, the specifying unit 150 executes a pixel specifying process. The pixel specifying process is a process of specifying, from a plurality of pixels included in the converted image 60, three types of pixels, that is, a changed background pixel, a maintenance background pixel, and an object-related pixel. The changed background pixel is a pixel constituting the above-described changed background image (in the example of FIG. 5A, the first background image 65 and the third background image 66). The maintenance background pixel is a pixel that constitutes the above-described maintenance background image (second background image 61 in the example of FIG. 5A). The object-related pixels are pixels constituting the above-described object image (images 62, 63, 64, and 67 in the example of FIG. 5A) and the boundary region MG adjacent to these object images.

  Specifically, the specifying unit 150 specifies a pixel in which at least one of three component values (RGB values) is 255 among a plurality of pixels constituting the converted image 60. The component value that is 255 of the identified pixel is a value that has been changed to 255 by the background color conversion process (changed value), or a value that is 255 (maintenance value) before the background color conversion process. The specifying unit 150 compares the original image 50 and the converted image 60 to determine whether the component value that is 255 of the specified pixel is a change value or a maintenance value. When the component value to be determined is a changed value, the specifying unit 150 specifies that the pixel including the component value is a changed background pixel. When the determination target component value is a maintenance value, the specification unit 150 specifies that the pixel including the component value is a maintenance background pixel. The specifying unit 150 specifies all the pixels excluding the changed background pixel and the maintenance background pixel among the plurality of pixels constituting the converted image 60 as the object-related pixels.

  In subsequent step S40, the specifying unit 150 executes a labeling process for attaching identifiers (numbers) to the three types of specified pixels. First, the identifying unit 150 labels “0” for the maintenance background pixel and “1” for the changed background pixel. Further, the specifying unit 150 groups a plurality of object-related pixels adjacent to each other, and assigns the same identifier (natural number of 2 or more (N + 1) or less: N is the number of groups) to each group. As a result, N processing regions are specified. The processing area is an area including the object image and the boundary area MG. For example, in the example of FIG. 5A, 13 processing areas respectively corresponding to 13 characters included in the character images 62, 63, 64 (SAMPLE, EFGH, WXY), and 1 corresponding to the photographic image 67. Two processing areas are identified.

  In subsequent step S50, the image processing unit 100 executes color correction processing. FIG. 7 is a flowchart of the color correction process. In step S505, the image processing unit 100 selects one processing area. In step S510, the determination unit 140 determines whether the selected processing region is adjacent to a number of changed background pixels equal to or greater than a threshold value. When the processing area is not adjacent to the number of changed background pixels equal to or greater than the threshold (step S510: NO), the image processing unit 100 proceeds to step S560. That is, color correction is not performed on the processing area to be processed. This is because, when the processing area is not adjacent to the number of changed background pixels equal to or greater than the threshold value, the unnaturalness of the boundary area MG due to the change in the color of the adjacent background image does not occur. This is because the necessity of targeting is low. For example, in the example illustrated in FIG. 5, the processing region including an image representing each character (EFGH) included in the second character image 63 of the converted image 60 is not a color correction target.

  When the processing region is adjacent to a number of changed background pixels equal to or greater than the threshold value (step S510: YES), the correction unit 160 of the image processing unit 100 calls a background color value (also referred to as an original background color value) before the background color conversion process. .) Is specified (step S515). Specifically, the correction unit 160 specifies a surrounding region surrounding the outer periphery of the processing region in the original image 50, and calculates an average value of pixel values of the surrounding region as an original background color value. The width of the surrounding area is a size corresponding to a predetermined number of pixels (for example, one pixel). The original background color value is represented by, for example, an RGB value and is represented as (Rbg, Gbg, Bbg).

In step S520, the correction unit 160 calculates the variance of the pixel values in the processing area. Here, the pixels used for calculating the variance are pixels (RGB background pixels) having a pixel value (RGB value) representing a color closer to the original background color value than the reference value from all the pixels constituting the processing region. This is a pixel excluding the pixel. The pixels to be excluded are pixels whose pixel values satisfy all the following expressions (1) to (3), for example.
Rbg−TH4 <R <Rbg + TH4 (1)
Gbg−TH4 <G <Gbg + TH4 (2)
Bbg−TH4 <B <Bbg + TH4 (3)

  Here, the variance to be calculated is the variance of the pixel values of the object image included in the processing area. Since the processing region includes the boundary region MG in addition to the object image, it is preferable to calculate the variance by excluding the pixels configuring the boundary region MG from the pixels configuring the processing region. The reason why the background approximate pixel is excluded from the pixels used when calculating the variance in this process is that the boundary region MG is considered to have a color that is relatively close to the original background color value. The variance calculated in this step is calculated for each of the three components, for example.

  In step S525, the determination unit 140 determines whether the processing region is substantially monochromatic using the calculated variance. For example, when all of the three variance values (σ_R, σ_G, σ_B) calculated for each component are equal to or less than the corresponding threshold values (TR, TG, TB), the determination unit 140 substantially determines the processing region. If it is determined to be a single color and at least one of the three variance values is greater than the corresponding threshold, it is determined that the processing region is not substantially a single color. Since pixel values (RGB values) of an image including a large number of values take values over a relatively wide range of possible ranges, at least one variance of variance values (σ_R, σ_G, σ_B) is relatively large. Become. On the other hand, the pixel value of a substantially monochromatic image is a relatively narrow range of possible ranges, and the dispersion values (σ_R, σ_G, σ_B) are relatively small. Therefore, if the threshold values (TR, TG, TB) are appropriately set, it can be appropriately determined whether the image is substantially a monochrome image. For example, generally, a photographic image 67 including a large number of colors is determined not to be a single color, and a single color character image such as a black character or a red character is determined to be a single color.

  When the processing region is not substantially monochromatic (step S525: NO), the correction unit 160 proceeds to step S560. That is, color correction is not performed on the processing area to be processed. When the processing area is substantially monochromatic (step S525: YES), the correction unit 160 converts the color system of the processing area from the RGB color system to the YCrCb color system (step S530).

  In step S540, the correction unit 160 identifies the color value of the object image included in the processing area. Specifically, the correction unit 160 calculates the average pixel value (YCC value) of the pixels used when calculating the variance described above, that is, the pixels excluding the background approximation pixels from all the pixels constituting the processing region. The color values of the object image (TC_Y, TC_Cb, TC_Cr) are calculated.

  In step S545, the correction unit 160 specifies a background color value (also referred to as a current background color value) after the background color conversion process. Specifically, the correction unit 160 specifies a surrounding region surrounding the outer periphery of the processing region in the converted image 60, and calculates an average value of pixel values of the surrounding region as the current background color value (BC_Y2, BC_Cb2, BC_Cr2). . The width of the surrounding area is a size corresponding to a predetermined number of pixels (for example, one pixel).

  In step S550, the correction unit 160 performs the color correction process using all pixels constituting the processing region as correction target pixels. In other words, the correction target pixel includes a plurality of pixels constituting the object image and a plurality of pixels constituting the boundary region MG. The correction unit 160 generates a color correction table using the color values (TC_Y, TC_Cb, TC_Cr) of the object image in the processing area and the current background color values (BC_Y2, BC_Cb2, BC_Cr2) of the surrounding background image. .

  8 and 9 are diagrams illustrating an example of the color correction table. FIG. 8 is a table used when the background color around the processing area is white ((BC_Y2, BC_Cb2, BC_Cr2) = (255, 0, 0)). FIG. 9 is a table used when the background color around the processing area is a color different from white. 8A and 9A show tables for correcting the chroma value Cb, and FIGS. 8B and 9B show tables for correcting another chroma value Cr.

  As shown in FIGS. 8 and 9, in these color correction tables, the luminance value Y is an input gradation value, and the chroma values Cb and Cr are output gradation values. The correction unit 160 refers to the color correction table using the luminance value Y of the correction target pixel, and acquires corrected chroma values Cb and Cr (correction values). The correction unit 160 changes the chroma values Cb and Cr of the correction target pixel to the chroma values Cb and Cr acquired as correction values. The correction unit 160 does not change the luminance value Y of the correction target pixel.

  As shown in FIGS. 8 and 9, the chroma values Cb and Cr after correction are set to the chroma values TC_Cb and TC_Cr of the object image in the range FR where the luminance value Y of the correction target pixel is lower than the luminance value TC_Y of the object image. Has been. Further, as shown in FIG. 9, the chroma values Cb and Cr after correction are the chroma values BC_Cb2 of the surrounding background image in the range HR where the luminance value Y of the correction target pixel is higher than the luminance value BC_Y2 of the current background color value. BC_Cr2 is set. As shown in FIG. 8, when the current background color value represents white (BC_Y2 = 255), there is no range HR.

  The corrected chroma values Cb and Cr are equal to the luminance value Y of the correction target pixel in the range GR where the luminance value Y of the correction target pixel is between the luminance value TC_Y of the object image and the luminance value BC_Y2 of the current background color value. Fluctuate accordingly. In other words, in the range GR, the corrected chroma values Cb and Cr become closer to the chroma values TC_Cb and TC_Cr of the object image as the luminance value Y of the correction target pixel is closer to the luminance value TC_Y of the object image. It is set so that the closer to the luminance value BC_Y2 of the color value, the closer to the chroma values BC_Cb2 and BC_Cr2 of the current background color value. For example, when the chroma value TC_Cb of the object image is larger than the chroma value BC_Cb2 of the current background color value (FIG. 8 (A), FIG. 9 (A)), the corrected chroma value Cb is the luminance in the range GR. As the value Y increases, it decreases monotonously. When the chroma value TC_Cr of the object image is smaller than the chroma value BC_Cr2 of the current background color value (FIG. 9B), the corrected chroma value Cr in the range GR corresponds to the increase of the luminance value Y. Increase monotonically.

  The result of such color correction will be described with reference to FIG. As described above with reference to FIG. 6A, the luminance value Y of the boundary region MG is closer to the luminance value TC_Y of the object image as it is closer to the position in contact with the object image (for example, the first position P1). It changes so that it is closer to the brightness value BC_Y2 of the background image as it is closer to the position (for example, the second position P2) in contact with the. Therefore, as a result of the color correction using the color correction table described above, the chroma value Cr of the boundary region MG after color correction is close to the position in contact with the object image, as shown by the one-dot broken line RC3 in FIG. 6B. The closer to the chroma value TC_Cr of the object image, the closer to the position in contact with the background image, the closer to the chroma value BC_Cr2 of the background image. As a result, even if an unnatural color is generated in the boundary region MG before correction, the unnatural color is corrected to a natural color by this color correction. In the color correction table range FR, the output gradation values (chroma values Cb, Cr) are set to a constant value, so that the dispersion of the chroma values Cb, Cr of the object image that is a single color is eliminated, and the single color is used. Color unevenness of an object image is reduced.

  In step S555, the correction unit 160 converts the color system of the processing area after the color correction process from the YCrCb color system to the RGB color system.

  In step S560, the image processing unit 100 determines whether all the processing areas have been selected in step S505. If the image processing unit 100 determines that all the processing areas are not selected (step S560: NO), the process returns to step S505, selects an unselected processing area, and repeats the above-described processing of steps S505 to S555. . When the image processing unit 100 determines that all the processing regions have been selected (step S560: YES), the image processing unit 100 ends the color correction processing (FIG. 7) and ends the image processing (FIG. 2).

  According to the present embodiment described above, an object feature value (specifically, YCC value) representing the color of the object image, a background feature value (specifically, YCC value) representing the color of the background image, Using the luminance value Y of the boundary region MG, the color of the boundary region MG is corrected so as to approach the color of the object image or the color of the background image. As a result, color unnaturalness that may occur in the boundary region MG can be suppressed. Therefore, the image quality of the boundary region MG and the image quality of the entire target image can be improved.

  Furthermore, according to the above embodiment, using the background feature values BC_Cr2, BC_Cb2 and the object feature values TC_Cr, TC_Cb, the color of the boundary region MG, the chroma value of the chroma value of the object image, and the background A color having a value between the chroma value of the image is corrected. As a result, color unnaturalness that may occur in the boundary region MG can be appropriately suppressed, and the image quality of the boundary region MG and the image quality of the entire target image can be improved.

  According to the above embodiment, correction is further performed on the color of the boundary region MG using the luminance value Y of the boundary region MG. Specifically, correction for changing the chroma values Cb and Cr according to the luminance value Y is executed. As a result, it is possible to appropriately suppress color unnaturalness that may occur in the boundary region.

  According to the above-described embodiment, the boundary region MG is further changed by changing the color value (specifically, chroma values Cb, Cr) different from the luminance value without changing the luminance value Y of the boundary region MG. Perform color correction. As a result, it is possible to maintain the gradation of the boundary region MG while suppressing the unnatural color of the boundary region MG. As a result, the boundary region MG can have a more natural image quality, and the image quality of the target image can be improved.

  According to the above-described embodiment, the luminance value Y of the correction target pixel is closer to the color value of the object image (specifically, the chroma values TC_Cb and TC_Cr) as the luminance value Y of the object image is closer. The color value of the boundary area MG is corrected. In addition, the color value of the boundary region MG is set so that the closer the luminance value Y of the correction target pixel is to the luminance value Y of the background image, the closer to the color value of the background image (specifically, chroma values BC_Cb2, BC_Cr2). to correct. Therefore, the color of the boundary region MG can be easily corrected to a color that is naturally connected to the object image and the background image.

  In the above embodiment, the determination unit 140 determines whether the object image to be processed is substantially monochromatic. Then, the correction unit 160 converts a plurality of pixels constituting a processing area including a single-color object image determined to be substantially monochromatic and a boundary area MG between a background area adjacent to the single-color object image, It is set as a correction target pixel. Although it may be difficult to accurately distinguish between the object image and the boundary region MG, in the above embodiment, the color of the boundary region MG is easily a natural color without distinguishing between the object image and the boundary region MG. Can be corrected.

  In particular, when the background color conversion process is executed as in the above embodiment, an unnatural color may occur in the boundary region MG as described above. According to the above embodiment, it is possible to improve the image quality of the boundary region MG by suppressing the unnaturalness of the boundary region MG of the image represented by the image data after the background color conversion process. Furthermore, the specifying unit 150 specifies the processing region by specifying the changed background pixel, the maintenance background pixel, and the object-related pixel. That is, a region corresponding to a plurality of object-related pixels adjacent to each other is highly likely to include an object image. Therefore, it is possible to easily and appropriately specify the processing area to be corrected.

  Further, among the processing areas, the area adjacent to the changed background pixel is subject to color correction, and the area adjacent to the maintenance background pixel is not subject to color correction. Can be corrected. As a result, the processing time and processing load of image processing can be reduced.

  Moreover, since the background color conversion process of the said Example is performed independently for every component, the color of the background image after conversion can change into various colors. Even in such a case, it is possible to improve the image quality of the boundary region MG by suppressing unnatural color that may occur in the boundary region MG of the image.

  Further, the color correction tables (FIGS. 8 and 9) of the above embodiment are set so that the chroma values Cb and Cr in the gradation value range FR smaller than the luminance value TC_Y of the object image are changed to constant values. Yes. As a result, the color correction process can suppress variations in hue and saturation occurring in an object image that is a single color. Therefore, not only the image quality of the boundary region MG but also the image quality of the single-color object image can be improved.

B. Variations:
(1) According to the above embodiment, the color of the boundary region MG is corrected to a color whose chroma value has a value between the chroma value of the object image and the chroma value of the background image. The color may be corrected to other colors. In this case, the corrected color is preferably an intermediate color between the color of the object image and the color of the background image. An intermediate color between the first color and the second color is a value of at least one component of the color system when the first color and the second color are represented by a specific type of color system. Is a value that is a value (also referred to as an intermediate value) between the value of the component of the first color and the value of the component of the second color. For example, when the cause of the unnatural color of the boundary region MG is the hue or saturation as in the above embodiment, the component values related to the hue and saturation are intermediate values in the employed color system. It is preferable to correct to an intermediate color. The component value related to the hue and saturation is generally a color system that includes a luminance value (a value that represents the brightness of the color, depending on the color system, including a value called brightness) as a component. The component value is different from the luminance value at. For example, the component values related to hue and saturation include Pb and Pr values in the YPbPr color system, a * values in the CIELAB color system, b *, in addition to the chroma values Cb and Cr in the YCrCb color system in the above-described embodiment. Value, hue (Hue) value in the HSB color system, saturation value, and the like. Further, an intermediate color between the first color and the second color, when the first color and the second color are represented by a specific type of color system, in the color space of the color system, It includes a point on a straight line connecting the position of the first color and the position of the second color.

  In the above embodiment, the luminance value Y of the boundary region MG is an intermediate value before correction (FIG. 6A), and the chroma values Cb and Cr are corrected to the intermediate value by the correction. As can be seen, it can be said that the color of the boundary region MG after correction is corrected to an intermediate color in which all three component values (Y, Cb, Cr) are intermediate values. Thus, it is preferable that the color of the boundary region MG after correction is corrected to an intermediate color in which a plurality of values among a plurality of component values in the adopted color system are intermediate values, and all the values are intermediate values. More preferably, the intermediate color is corrected.

(2) According to the above embodiment, the color of the boundary region MG is corrected so that the chroma value approaches the chroma value of the object image or the background image. However, the color of the boundary region MG is not limited to this. What is necessary is just to correct | amend so that the color of an object image and the color of a background image may approach at least one color. In order to correct the color of the boundary region MG so as to approach at least one of the color of the object image and the color of the background image, in the above-described embodiment, the characteristic value relating to the color of the object image (in the embodiment, YCC Value), a background feature value representing the color of the background image (specifically, a YCC value), and a luminance value Y of the boundary region MG, all three values need not be used. For example, the boundary region MG may be corrected using a combination of the feature value related to the color of the object image and the luminance value Y of the boundary region MG, or the feature value related to the color of the background image and the boundary region MG The boundary region MG may be corrected using a combination of the luminance value Y of MG. For example, when the object image is a character image, if the color of the boundary region MG is close to the color of the character image, the character appears to be thick, and if it is close to the color of the background image, the character tends to appear thin. The color of the background image or the color of the character image may be approximated according to the character mode or the user's preference. In this case, the boundary region MG is corrected using the feature value and the luminance value Y of the color of the background image and the color of the character image that are closer to the color of the boundary region MG by correction. May be.

  Here, correcting the first color so as to approach the second color means that when the first color and the second color are represented by a specific type of color system, at least one of the first color It includes changing the component value of the first color so that the difference between the value of one component and the value of the component of the second color is smaller than before correction. In this case, it is preferable to change the component value of the first color so that there is no component in which the difference is larger than that before the correction. Further, the correction of the first color so as to approach the second color means that the first color and the second color are expressed in a specific color system in the color space of the color system. And changing the component value of the first color so that the Euclidean distance between the first color and the second color is shortened.

(3) Unlike the above embodiment, the object image is represented by a relatively light color (color in which each RGB value is relatively large), and the background image is a relatively dark color (each RGB value is Image data representing an image (also referred to as a grayscale inverted image) represented by a relatively small value) may be assumed as target image data. In this case, the background color conversion process is, for example, a process of converting a relatively dark color to black ((R, G, B) = (0, 0, 0)). In this case, for example, instead of the color conversion table RT shown in FIG. 4, the background color conversion process is performed for the input gradation value Vin within a predetermined range (for example, 0 ≦ Vin <THa) on the low gradation value side. Corresponding to the change, all the output gradation values Vout are changed to the minimum gradation value (0), and corresponding to the change of the input gradation value Vin within the predetermined range (THb <Vin ≦ 255) on the high gradation value side. Thus, the color conversion table is set so as to maintain the output gradation value Vout at the same gradation value. In this case, in the color conversion table, the output gradation value Vout is set so that the output gradation value Vout does not become discontinuous corresponding to the change in the input gradation value Vin within (THa ≦ Vin ≦ THb). It may be set.

  For color correction on the grayscale inverted image, instead of the color correction table shown in FIGS. 8 and 9, for example, the luminance Y of the boundary region MG is the luminance value TC_Y of the object image (the luminance value “255” corresponding to white) If a larger value is assumed), a color correction table is used in which the chroma value Cr of the correction target pixel becomes the same value as the chroma values TC_Cr and TC_Cb of the object image. This color correction table is further corrected when the luminance value Y of the boundary region MG is smaller than the luminance value BC_Y2 of the background image (a luminance value “0” corresponding to black or a relatively small value is assumed). The chroma values Cr and Cb of the target pixel are set so as to be the same values as the chroma values BC_Cr2 and BC_Cb2 of the background image. Further, in this color correction table, when the luminance value Y of the boundary region MG is within the range of BC_Y2 ≦ Y ≦ TC_Y, the luminance value Y of the correction target pixel is close to the luminance value TC_Y of the object image. The closer to the chroma values TC_Cb and TC_Cr of the object image, the closer to the luminance value BC_Y2 of the background image, the closer to the chroma values BC_Cb2 and BC_Cr2 of the background image. As a result, an unnatural color that can appear in the boundary region MG of the grayscale inverted image is eliminated by correction, and the image quality of the grayscale inverted image can be improved.

(4) In the above embodiment, the average value of the YCC values of the pixels constituting the object image is used as the feature value relating to the color of the object image. However, the present invention is not limited to this, and the median value and the mode value are used. It may be used, or a statistical value (average value, median value, mode value, etc.) calculated using all or part of the component values when expressed in other color systems may be used.

(5) The image processing of the above embodiment includes the background color conversion process, but may not include the background color conversion process. The image processing of the above embodiment is not limited to the multi-function device 200, and may be executed by other devices such as a personal computer or a server. For example, scan data that has undergone background color conversion processing may be acquired from a scanner, and a personal computer may execute image processing that does not include background color conversion processing using the scan data as target image data.

(6) In the above embodiment, among the background pixels, a pixel having a pixel value that has not been changed before and after the background color conversion process is specified as a maintenance background pixel, and a pixel having the changed pixel value is changed to the changed background pixel. As specified. Not limited to this, a pixel having a pixel value whose change amount before and after the background color conversion process is 5 or less is specified as a maintenance background pixel, and a pixel having a pixel value of 6 or more is specified as a change background pixel You may do it. Generally speaking, a pixel having a pixel value whose pixel value before the background color conversion process and the pixel value after the background color conversion process is less than or equal to a reference value is identified as a maintenance background pixel, A pixel having a pixel value whose difference is larger than the reference value may be specified as the changed background pixel. In this case, even if it is a substantially white background image due to unevenness or the like, and even if the pixel value can be changed by the background color conversion process, the maintenance background pixel, the changed background pixel, , Can be specified.

(7) The background color conversion process in the above embodiment is executed independently for each component, but is not limited thereto. For example, all the RGB component values are changed to white when the pixel is within the change range CR (FIG. 4), and at least one component value of RGB is outside the change range CR. The value may not be changed). In this case, since the color of the background image whose color has been changed by the background color conversion process is limited to white, the specification of the background color value in step S545 in FIG. 7 can be omitted. Further, the color of the pixel value to be changed by the background color conversion process is not limited to white, but may be another color (for example, light yellow). Even in this case, an unnatural color in the boundary region MG can be corrected appropriately by using an appropriate color correction table (FIG. 9) according to the background color after the change.

(8) In the above embodiment, a part of the configuration realized by hardware may be replaced with software. Conversely, a part of the configuration realized by software is replaced with hardware. Also good.

  As mentioned above, although this invention was demonstrated based on the Example and the modification, Embodiment mentioned above is for making an understanding of this invention easy, and does not limit this invention. The present invention can be changed and improved without departing from the spirit and scope of the claims, and equivalents thereof are included in the present invention.

  DESCRIPTION OF SYMBOLS 100 ... Image processing part, 200 ... Multifunction machine, 210 ... CPU, 222 ... Computer program, 220 ... Nonvolatile storage device, 230 ... Volatile storage device, 231 ... Buffer area 240 ... Printer section 250 ... Scanner section 260 ... Operation section 270 ... Display section 280 ... Communication section 300 ... Digital camera 400 ... Personal Computer,

Claims (12)

An image data acquisition unit for acquiring target image data representing a target image including an object image and a background image adjacent to the object image;
A feature value acquisition unit that acquires a first feature value relating to the color of the object image and a second feature value relating to the color of the background image;
Using the first feature value and the second feature value, the color of the boundary area, which is an area where the object image and the background image are adjacent, is set to the color of the object image and the color of the background image. A correction unit that corrects an intermediate color indicating a color between
An image processing apparatus comprising: The image processing apparatus according to claim 1,
The correction unit further performs the correction on the color of the boundary region using the luminance value of the boundary region. An image data acquisition unit for acquiring target image data representing a target image including an object image and a background image adjacent to the object image;
A feature value acquisition unit that acquires at least one of the first feature value related to the color of the object image and the second feature value related to the color of the background image;
Using at least one of the acquired feature values of the first feature value and the second feature value, and a brightness value of a boundary region that is a region where the object image and the background image are adjacent to each other, A correction unit that corrects the color of the boundary region so as to approach at least one of the color of the object image and the color of the background image;
An image processing apparatus comprising: The image processing apparatus according to claim 1 or 3, wherein
The correction unit changes a color value that represents a color of the boundary region and is different from the luminance value without changing a luminance value that represents the luminance of the boundary region. An image processing apparatus that performs the correction on the color of the boundary region. The image processing apparatus according to claim 4,
The first feature value includes a first luminance value that represents the luminance of the object image and a first color value that is the color value that represents the color of the object image,
The second feature value includes a second luminance value representing the luminance of the background image and a second color value which is the color value representing the color of the background image,
The correction unit performs the correction on the color of the boundary region by changing the color value of the correction target pixel including the pixel of the boundary region to a correction value,
The correction value is such that when the correction target pixel has a luminance value between the first luminance value and the second luminance value, the luminance value of the correction target pixel becomes the first luminance value. Depending on the luminance value of the correction target pixel, the closer to the first color value, the closer the luminance value of the correction target pixel to the second luminance value, the closer to the second color value. An image processing device set by The image processing apparatus according to claim 5, further comprising:
A determination unit for determining whether the object image is substantially monochromatic;
The correction unit includes a plurality of pixels constituting a region including a single-color object image that is the object image determined to be substantially a single color, and the boundary region between the single-color object image and the background region. An image processing apparatus as the correction target pixel. The image processing apparatus according to any one of claims 1 to 6, further comprising:
A conversion processing unit that performs color conversion processing on the target image data to convert the color of the background image, and the color conversion processing is processing for changing a pixel value within a specific range to a specific value. Including the conversion unit including:
The image processing apparatus, wherein the second feature value relating to the color of the background image is a feature value relating to the color represented by the specific value. The image processing apparatus according to claim 7, further comprising:
Among the plurality of pixels constituting the target image data that has been subjected to the color conversion process, the specifying unit is configured to identify a first type pixel and a second type pixel, and the first type pixel is , The pixel having the specific value, and the second type pixel includes the specific unit, the pixel having a pixel value different from the specific value,
The first feature value related to the color of the object image is a feature value related to a color expressed using a plurality of pixel values of a plurality of second type pixels adjacent to each other. The image processing apparatus according to claim 8,
The first type of pixel includes a first pixel in which a difference between a pixel value before the color conversion process and a pixel value after the color conversion process is equal to or less than a reference value, and a first pixel in which the difference is greater than a reference value. 2 pixels,
The specifying unit specifies the first pixel and the second pixel, respectively.
The feature value acquisition unit obtains the first feature value and the second feature value when an area corresponding to the plurality of second type pixels adjacent to each other is adjacent to the second pixel. An image processing apparatus that acquires at least one feature value. An image processing apparatus according to any one of claims 7 to 9,
The target image data includes a plurality of component image data corresponding to each of a plurality of components constituting the color system,

The color conversion process is a plurality of component conversion processes executed independently for each of the plurality of component image data, and pixel values in a component specific range that is the specific range of the component to be processed are calculated. An image processing apparatus including the plurality of component conversion processes for changing to a component specific value that is the specific value of the component to be processed. An image data acquisition function for acquiring target image data representing a target image including an object image and a background image adjacent to the object image;
A feature value acquisition function for acquiring a first feature value relating to the color of the object image and a second feature value relating to the color of the background image;
Using the first feature value and the second feature value, the color of the boundary area, which is an area where the object image and the background image are adjacent, is set to the color of the object image and the color of the background image. A correction function for correcting to an intermediate color indicating a color between
A computer program that causes a computer to realize An image data acquisition function for acquiring target image data representing a target image including an object image and a background image adjacent to the object image;
A feature value acquisition function for acquiring at least one feature value of a first feature value relating to the color of the object image and a second feature value relating to the color of the background image;
Using at least one of the acquired feature values of the first feature value and the second feature value, and a brightness value of a boundary region that is a region where the object image and the background image are adjacent to each other, A correction function for correcting the color of the boundary region so as to approach at least one of the color of the object image and the color of the background image;
A computer program that causes a computer to realize

Images